To demonstrate physiologic changes associated with asthma symptoms that many patients with asthma develop during exercise, we used sustained constant-load and interval exercise protocols with subjects breathing dry room temperature air. In constant-load exercise, subjects pedaled a stationary bicycle at 50% of their maximal power capacity for 36 min. In interval protocols, subjects pedaled at 60% of maximal capacity for 6 min and then 40% of maximal for 6 min; the 12-min cycle was repeated three times for a total exercise time of 36 min. Maximal expiratory flow versus volume maneuvers (MEFV) were obtained before, at 6-min intervals during, and at 5-min intervals after exercise. Changes in peak expiratory flow (PEF), forced expiratory volume in 1 s (FEV(1)), and forced expiratory flow at 50% of pre-exercise vital capacity (FEF(50)) were compared with pre-exercise values. Within 15 min after a maximal 1-min incremental exercise protocol, mean flows decreased compared with pre-exercise (PEF, mean -22%, range -46 to 5%; FEV(1), mean -21%, range -42 to -3%; FEF(50), mean -41%, range -80 to 3%; all p < 0.05). There were no significant changes in MEFV flows until 18 min of constant-load exercise, when FEV(1) and FEF(50) fell (FEV(1), mean -6%, range -15 to 2%; FEF(50), mean -14%, range -32 to 6%; both p < 0.05), although changes in PEF were minimal and were not significantly different compared with pre-exercise. During the interval protocol, mean flows declined each time work load was reduced from 60 to 40% of maximal capacity (PEF, -10%, range -26 to 7%; FEV(1), mean -10%, range -22 to -3%; FEF(50), mean -24%, range -55 to 3%; all p < 0.05), and flows increased toward pre-exercise values each time work load was increased again to 60% of maximal capacity. We conclude that in subjects with EIA, bronchoconstriction can occur during exercise, particularly during periods of reduced exercise intensity, although the constriction is reversible by a return to higher exercise intensity. There is little refractoriness to repeat cycles of periods of high followed by low work rates. Airway function during and after exercise likely reflects a dynamic balance between bronchoconstrictor and bronchodilator influences.